Can I send a message to all agents at once?

Yes, using the 'talk all' command allows you to broadcast a message to every active agent in your tmux-team configuration simultaneously.

Which AI models can I communicate with?

You can communicate with any agent configured in your tmux environment, including Codex, Gemini, or other Claude instances.

Why should I use the --wait flag?

The --wait flag is more token-efficient because it blocks until a response is received, removing the need for the primary agent to repeatedly poll for status updates.

How do I see which agents are available?

You can run the 'tmux-team list' command to see a comprehensive list of all configured agents available for collaboration.

Tmux Team is a coordination skill for Claude Code that allows it to communicate with and delegate tasks to other AI agents running in different tmux panes.

Tmux Team

Name: Tmux Team
Author: wkh237

bywkh237

•

Collaboration Tools

Enables seamless coordination and task delegation between multiple AI agents running in separate tmux panes.

About

Tmux Team is a specialized coordination skill designed for multi-agent workflows within a tmux environment. It allows Claude to act as a project lead, delegating specific tasks, requesting code reviews, or broadcasting updates to other AI instances like Codex or Gemini. By utilizing a unified CLI, this skill facilitates parallel processing and cross-model collaboration, making it ideal for complex engineering projects that require specialized input or concurrent execution across multiple terminal sessions.

Key Features

Synchronous messaging with the --wait flag for optimized token utilization
1 GitHub stars
Support for multi-model collaboration including Codex, Gemini, and Claude
Direct agent-to-agent communication via the talk command
Global broadcast capabilities to message all active agents simultaneously
Customizable timeouts and delays for managing long-running agent tasks

Use Cases

Requesting peer reviews from different AI models to gather diverse architectural perspectives
Coordinating parallel development workflows across multiple terminal panes to increase throughput
Delegating specialized sub-tasks like unit test generation or security audits to secondary agents

About

Key Features

Synchronous messaging with the --wait flag for optimized token utilization
1 GitHub stars
Support for multi-model collaboration including Codex, Gemini, and Claude
Direct agent-to-agent communication via the talk command
Global broadcast capabilities to message all active agents simultaneously
Customizable timeouts and delays for managing long-running agent tasks

Use Cases

Requesting peer reviews from different AI models to gather diverse architectural perspectives
Coordinating parallel development workflows across multiple terminal panes to increase throughput
Delegating specialized sub-tasks like unit test generation or security audits to secondary agents